Abstract Scope |
Co-Cr-Mo has been employed for production of permanent orthopedic implants. However, when implanted it is prone to suffer from tribocorrosion, releasing Co and Cr ions, causing adverse reactions for human health. The ball burnishing technique has been successfully employed to improve the surface integrity in metal alloys, including Co-Cr-Mo, however, optimization of the process needs to be carried out to reduce energy and time consumption. In this work we used the artificial neural networks (ANN) to find an empirical relationship between burnishing force and number of passes, and roughness, hardness and corrosion resistance of Co-Cr-Mo. A 32 factorial design of experiments was used for ANN model development. The number of nodes of the input, hidden and output layers was 2, 7 and 3, respectively. Results showed that maximum errors between predicted and experimental values are less than 10%. Finally, the surface properties where optimized through the NSGA method. |